Finding the Fremington Clay Erratics - Updated

Finding the Fremington Clay Erratics

 I have been studying the original descriptions of the erratics found in the Fremington Clays, which are very different to the nearby Coastal Ice Rafted erratics, and sadly they have been lumped in with them which has caused confusion.

I think I have found two of the original half a dozen, so they could be examined again and the very old descriptions of them could be updated.

The original references are:

Dewey, H. (1910). Notes on some igneous rocks from North Devon. Proceedings of the Geologists' Association, 21(4), 429–434.

Taylor, C.W. (1956) Erratics of the Saunton and Fremington areas. Report and Transactions of the Devonshire Associaton for the Advancement of Science, Literature and Art, 88, 52–64.)

  The following  photos and text are from Taylor (1956):

 

 

There are two boulders from this district which are mentioned in the literature, and both of them were found without any difficulty, as they are preserved, and it would appear that enquiries for them are not infrequent. The present owner of Combrew Farm, Mr. Tucker, is aware of their interest, and unnecessary "hammering" is discouraged. The third boulder, which, though mentioned, has not so far been described or figured, came originally from the clay-pit of the old Fishley Pottery, now long disused, while as regards recent finds, two further masses have been obtained from the brown clay of Brannam's pit and are noted below. Although the literature makes no mention of the fact, the boulders Nos. 6 and 7 following are understood from local information to have been found also in the clay of the Fishley pit.

No. 6. A boulder is mentioned by Dewey (“a dark grey, finely crystalline rock with small white porphyritic felspars in the groundmass”)  as at Bickington, about a quarter mile east of Combrew Farm, but owing to development, it has recently been moved to the garden on the right of the gateway to the main yard of the Farm. In case any ultra-zealous Glacialist should fit certain scars on portions of this rock, the author hastens to mention that it was dragged in its present position instead of being ruthlessly broken up on site, a consideration which is highly appreciated.

The original measurements and photograph have been given by Dewey, which need not be repeated here; it may be added, however, that the mass shows no distinct striae or wedge-shape, and the boulder measures roughly 40 x 30 x 25 inches high, the longest axis pointing almost E-W. It may be a Cornish spilite, of the pillow-lava type of igneous rock, which occurs in the Meneage district and elsewhere in Cornwall.

No. 7. The next erratic of this group is the hyalopilitic andesite, also previously described with No. 6 above. It is now situated on the right of the gated portion of the driveway to Combrew Farm, and is a glassy, brittle andesite, quite different from any of the foregoing rocks. Well rounded and about sixteen inches across, it contains no augite, but otherwise resembles similar rocks of Dumfries and Loch Craignish, Argyllshire.

No. 8. The rock from the Fishley Pottery clay-pit has hitherto merely been mentioned as "an igneous boulder" which may now be found on the right-hand, inside the gate of the first building (the old disused pottery) on approaching Combrew Farm from the main road. It is well covered with lichen, with flat top and base, and measures 47 x 19 x 16 inches high. Of a light grey colour, it is holocrystalline, with some quartz and much felspar, which appears altered; little mica could be observed. The texture is rather granitic, with a fine, pale coloured base containing the larger phenocrysts of quartz and felspar up to 5 mm. in size, but averaging 3 mm.

Various rocks, maybe the erratics, appear to still be on the roadside and pictured on Google Streetview.

  https://maps.app.goo.gl/VWN42wA1u1PbG6zZ9

 https://maps.app.goo.gl/ijcV53LLrDUywAKR8

 

 https://maps.app.goo.gl/mMB82AMYzt4KxPjN7

 https://maps.app.goo.gl/8uvFckNu1TB7BFhd6

 

 https://maps.app.goo.gl/iocQpaVFEwAGqQ176

 

UPDATE – I visited the farm on 2 November 2025 and photographed and measured the rocks.

 

No.7, the 16 inch well rounded erratic, which was on the wall, and was given a possible Scottish source is still there. This is the most important erratic in determining the source of the clays and erratics.

  

 

 

Erratic No.6 may be also by the farm entrance, the size is similar to the quoted size (40 x 30 x 25 inches) and it is “a dark grey, finely crystalline rock” .

If it is the same erratic it has been rotated so it doesn’t match Taylor’s photo exactly.

  

 

The other two roadside erratics are much smaller and are conglomerates, one looks very like a sarsen type puddingstone. They don’t match any of the erratic descriptions.

The 1938 Stonehenge Vandalism Incident: Larkhill Officers and the Onset of War

On the night of 16 June 1938, four recently commissioned Second Lieutenants of the Royal Artillery, William Laurence Sherrard, William Howard Skinner, John Edward Passingham Pierce, and John Lambert Shearme (erroneously noted as Shearne in some reports) engaged in an act of vandalism at Stonehenge. This episode, occurring amid the escalating tensions preceding the Second World War, exemplifies the occasional lapses in discipline among junior officers during a period of transition from peacetime routines to wartime preparedness. All four were in their early twenties, having completed training courses at Larkhill Garrison near Salisbury, and were due to depart for new postings the following day.

The Incident

Following a guest night at the Larkhill officers' mess, the group acquired green paint and brushes from the garrison's tennis courts. They applied paint to four upright sarsen stones in the main circle and portions of the Heel Stone also know as the Friar's Heel. Reports indicate the inscription of the phrase "Does this look like a friar?" on the stone. Additionally, several chamber pots were placed atop the affected monoliths.

They also altered a road sign on the London-Exeter trunk road (the A303). Contemporary accounts describe the addition of a letter to the destination "Exeter," though the precise modification—believed to render it as "Sexeter"—was omitted from published reports due to its indelicate nature.

The vandalism was discovered the following morning, prompting immediate concern from the site's custodians and local authorities.

Legal Proceedings and Response

The incident drew widespread press attention, with an estimated 60 officers from Larkhill initially requested to come forward. Sherrard, Skinner, Pierce, and Shearme voluntarily confessed, demonstrating a commitment to accountability consistent with the era's expectations of officers and gentlemen. They appeared before Salisbury Magistrates' Court, where they pleaded guilty to charges of criminal damage.

Each was fined £1, with the group collectively ordered to pay £20 in costs and restoration expenses. The presiding magistrate noted that the stones' natural patina might require up to 1,000 years of weathering to fully recover. Their commanding officer at Larkhill issued an official reprimand, but no further military sanctions were recorded. The leniency reflects the context of youthful indiscretion on the cusp of war, when such pranks were not uncommon among subalterns.

Wartime and Postwar Careers

The outbreak of war in September 1939 dispersed the group, their subsequent records reflecting diverse trajectories within the Royal Artillery:

• William Laurence Sherrard: Commissioned on 28 January 1938 after training at the School of Anti-Aircraft Defence, Biggin Hill. Promoted to Captain, he deployed to Sumatra in 1941 to defend oil installations. Killed in action on 14 February 1942 during a Japanese paratroop assault at Pladjoe airfield, where his unit faced superior numbers. Mentioned in Despatches for gallantry; commemorated on the Ditchling War Memorial, Sussex.
• William Howard Skinner: Born in Bengal in 1918. Also commissioned on 28 January 1938, following anti-aircraft training at Biggin Hill. No further public records of promotions, postings, or honours have been identified, though his survival through the war is confirmed by the absence of casualty entries. Likely served in defensive roles during the early conflict. Died Lingfield, Surrey 2000.
• John Edward Passingham Pierce: Commissioned on 28 January 1938, attached to the 22nd Anti-Aircraft Battery at the Royal Artillery Experimental Camp, Watchet. Advanced to Lieutenant-Colonel (service number 74538) by the early 1960s. Appointed Member of the Order of the British Empire (MBE) in the 1962 New Year Honours for services to the Royal Artillery; retired to the Regular Army Reserve of Officers thereafter.
• John Lambert Shearme: Commissioned on 28 January 1938, attached to the Coast Artillery School at Shoeburyness. Promoted to Major (service number 74525); retired on pay on 29 September 1958, retaining reserve liability. His coastal defence specialisation suggests involvement in home-based anti-invasion measures.

Fremington Clay Erratics - Dewey and Taylor descriptions

 From Dewey, H. (1910). Notes on some igneous rocks from North Devon. Proceedings of the Geologists' Association, 21(4), 429–434.

 

…boulders (III. and IV.) occur in the neighbourhood of Bickington, one (No. IV.) in front of Combrew Farm, and the other (No. III.) on a green about a quarter of a mile to the east of this locality.

The boulder No. III. (green quarter of a mile east of Combrew, 6-inch Ordnance Map, Devon, xiii, N.W.; 1-inch Sheet 293 Barnstaple) is a dark grey, finely crystalline rock with small white porphyritic felspars in the groundmass. Under the microscope it is seen to be rich in albite felspar, which occurs in the form of small plates and laths (Pl. XXIX, Fig. 3). These felspars lie in a groundmass of micropegmatite, but no free quartz is visible. In parts of the slide the granophyric groundmass appears to fill vesicles, and in other cases vesicles are filled with pleochroic calcite. If the rock ever contained a ferromagnesian constituent it could not have been plentiful, and is now completely replaced by chlorite.

In many respects the rock resembles the spilites of North and East Cornwall. The granophyric structure is not seen in the spilites, and in this rock it is mostly of secondary origin filling the vesicles. The other characters, however, resemble the spilites very closely, namely, the small albite felspars, the scarcity of ferromagnesian minerals and their replacement by chlorite; the absence of free quartz and the vesicular groundmass. It is possible that this boulder was derived from North Cornwall, but it is not safe to infer from the data available that it was probably derived from that area.

(click to enlarge)

By far the most interesting rock of those collected is the one found in front of Combrew Farm, boulder No. IV (6-inch Ordnance Map, Devon, xiii, N. W.; 1-inch Sheet 293, Barnstaple). It has been placed on the wall bounding a flower garden in front of the house. The rock is dark grey-green, with large pale olive-green felspars which are glassy and easily chipped from the rock. Microscopic examination reveals its glassy and porphyritic nature (Pl. XXIX., Fig. 4). It possesses felspars of two generations, but both are acid labradorite. The larger ones form a quarter of the rock, and the smaller occur in about equal quantities with the glassy base. The ferromagnesian constituent is a rhombic pyroxene which occurs to the entire exclusion of all other ferromagnesian minerals, for there is no augite, hornblende, or olivine. It occurs as small prisms, nearly the same size as the small felspars of the ground mass, with good cleavages, strong pleochroism, and straight extinction, and the pleochroism ranges from green to reddish yellow. Of the felspathic constituents the larger crystals are perfectly fresh and twinned on the albite and pericline laws, have high angles of extinction in symmetrical twins on each side of the plane of twinning, high refractive indices (the refractive index of this felspar is 1560, and the felspar is thus composed of 50% Ab. and 50% An., or in other words, is an acid labradorite), and inclusions arranged in well-marked zones. The small felspars are twinned on the albite law, but do not differ from the larger ones in their composition and freshness. There is no quartz present in the slide, but magnetite is abundant, and occurs as rods and feathery masses, and also as fine thin lines in the glassy base forming a network or gridiron structure. The crystalline constituents are embedded in a brown glass which constitutes about half the rock. In places this glass passes into the large felspar crystals by insensible gradations, there being no sign of a crystalline boundary between the two; but there is a well-marked zonal border in the brown glass surrounding the felspar.

The rock may be described as a hypersthene andesite. In many respects it resembles the Tholeite of Watt Carrick, Dumfries (1-inch Sheet 10, Scotland), and the hypersthene rocks of Curachan, Loch Craignish, Argyll (1-inch Sheet 36, Scotland), but all of these rocks contain considerable quantities of augite, whereas this rock is free from augite.

 

--------------------------------------------------------------------------------------------

 

From: Taylor, C. W., 1956. Erratics of the Saunton and Fremington areas. Rep. Trans. Devon. Ass. Advmt Sci., 88, 52—64.

 Fremington Area

"There are two boulders from this district which are mentioned in the literature, and both of them were found without any difficulty, as they are preserved, and it would appear that enquiries for them are not infrequent. The present owner of Combrew Farm, Mr. Tucker, is aware of their interest, and unnecessary "hammering" is discouraged. The third boulder, which, though mentioned, has not so far been described or figured, came originally from the clay-pit of the old Fishley Pottery, now long disused, while as regards recent finds, two further masses have been obtained from the brown clay of Brannam's pit and are noted below. Although the literature makes no mention of the fact, the boulders Nos. 6 and 7 following are understood from local information to have been found also in the clay of the Fishley pit.

No. 6. A boulder is mentioned by Dewey¹¹ as at Bickington, about a quarter mile east of Combrew Farm, but owing to development, it has recently been moved to the garden on the right of the gateway to the main yard of the Farm. In case any ultra-zealous Glacialist should fit certain scars on portions of this rock, the author hastens to mention that it was dragged in its present position instead of being ruthlessly broken up on site, a consideration which is highly appreciated.

The original measurements and photograph have been given by Dewey, which need not be repeated here; it may be added, however, that the mass shows no distinct striae or wedge-shape, and the boulder measures roughly 40 x 30 x 25 inches high, the longest axis pointing almost E-W. It may be a Cornish spilite, of the pillow-lava type of igneous rock, which occurs in the Meneage district and elsewhere in Cornwall.

No. 7. The next erratic of this group is the hyalopilitic andesite, also previously described with No. 6 above. It is now situated on the right of the gated portion of the driveway to Combrew Farm, and is a glassy, brittle andesite, quite different from any of the foregoing rocks. Well rounded and about sixteen inches across, it contains no augite, but otherwise resembles similar rocks of Dumfries and Loch Craignish, Argyllshire.

No. 8. The rock from the Fishley Pottery clay-pit has hitherto merely been mentioned as "an igneous boulder" which may now be found on the right-hand, inside the gate of the first building (the old disused pottery) on approaching Combrew Farm from the main road. It is well covered with lichen, with flat top and base, and measures 47 x 19 x 16 inches high. Of a light grey colour, it is holocrystalline, with some quartz and much felspar, which appears altered; little mica could be observed. The texture is rather granitic, with a fine, pale coloured base containing the larger phenocrysts of quartz and felspar up to 5 mm. in size, but averaging 3 mm.

A section of this specimen under the microscope shews it to be a much altered quartz porphyry, with crushed and irregular crystals of plagioclase, porphyritic quartz and long prisms of apatite, with a matrix nearly amorphous and red; epidote appears to have replaced part of the mosaic. This is regarded as a highly peculiar textural type, which may be derived from a fairly local source, such as the porphyritic dyke, west of the coasts of Devon and Cornwall.

No. 9. Across the fields in a south easterly direction from Combrew Farm, and through the coarse white clay in the adjacent Brannam's pits (to be distinguished from the Fishley Pottery) there were found two masses of boulders. The better-shaped boulder of the pair showed considerable rounding and exhibits a regular ellipsoidal form, but with lower mean weight, which was found in the clay. It is a highly crystalline and compact igneous rock of a grey colour, the texture resembling a very fine grained granite; a soft milky white crystal of felspar can be seen with a lens, while weak and indicates the presence of a little calcite. Also it includes highly kaolinized felspar, but still revealing the original lath outline, with a small proportion of quartz. Augite is reddish fresh, altering only slightly around the edges, while a little magnetite occurs, and a small amount of secondary calcite. Long needle prisms of apatite are present, with longitudinal inclusions and penetrating other minerals. The quartz, to judge from the textural arrangement, is probably of primary formation.

This rock appears to be a quartz dolerite, with none but a trace of a sub-ophitic texture, and has no characteristic which would enable a source to be named. Its main interest, as is the case with the other boulders here, lies in being included amongst this deposit of clay, the origin of which has given rise to intermittent discussion since the 1860's.

No. 10. A further rock from the same clay-pit as No. 9 may be found, in two pieces, by the surface at the far side of the pit, also of about three hundredweight, and of irregular, angular shape. It is of somewhat similar appearance to the foregoing, but of a darker grey and more crystalline. Again soft felspar crystals may be observed in the interior and with some calcite, indicated by weak acid. Examination of a thin section shows micro-pegmatitic ophitic structure with crossed flows, with tabs of plagioclase enclosed, and prisms of ilmenite are oriented transverse to the cracks. Grains of olivine under the thin section appear yellow. The quartz is light in colour but the texture shows slight orientation. Some extinction occurs here and there in the felspars. This boulder, an olivine dolerite, is quite fresh, but like the foregoing example, offers no means of determining its source; as mentioned under 5b, it is not an uncommon type in Devon.

Through the kind advice of Miss M. A. Arber, the Brannam clay-pit was later again visited to inspect further smaller finds which had been obtained at a depth of 17 feet. One of these proved to be a two-inch smoothed and rounded pebble of olivine dolerite, similar to No. 10 above, while the other comprised a five-inch slab of waterworn Carboniferous grit an inch and a quarter thick. This fragment has its two flat surfaces still partly covered with a smooth, red skin which proved to be ferric oxide, probably deposited by infiltration in situ along bedding cracks. These pebbles call for no further comment, save perhaps that the skin suggests that the grit was waterworn mainly before inclusion in the clay, as otherwise the skin might be expected to be completely removed.

Thus, so far as can be traced from the literature at the author's disposal, all the principal erratics mentioned from time to time as being in this area are still in existence, while a few additional blocks of local or foreign sources may be found at odd points. The older literature also mentioned casually by J. Prestwich, "a smaller block of fine grained white granite, and another of quartz grit, on the beach, S. of the cliff." Neither of these is known to the present writer, while there are several blocks of igneous rock to be found seen on the river or covered by sand. Again in his 1879 Memoir of the North Devon Athenæum there is a specimen labelled "Granite Boulder, Raised Beach, Baggy", but no other note is available.

---

 Summary of Fremington Erratics: Dewey (1910) and Taylor (1956)

Original Descriptions

Dewey (1910)

Two principal igneous boulders were described near Combrew Farm, Bickington:

1. Boulder III (east of Combrew Farm)
• Fine-grained dark grey rock with small white felspars (albite).
• Groundmass of micropegmatite, vesicular with calcite fillings.
• No quartz, little or no ferromagnesian minerals (replaced by chlorite).
• Compared to spilites (altered basalts) of North and East Cornwall, though not identical.
• Possible but uncertain source: North Cornwall.
2. Boulder IV (Combrew Farm garden)
• Dark grey-green, glassy and porphyritic, with large acid labradorite felspars.
• Contains hypersthene as only ferromagnesian mineral (no augite or olivine).
• Abundant magnetite; about half of rock is brown glass.
• Identified as a hypersthene andesite.
• Comparison: Similar to rocks from Watt Carrick (Dumfries) and Loch Craignish (Argyll), but differing by lack of augite.
• Suggested source: Uncertain—Scotland suggested by comparison, but not confirmed.

---

Taylor (1956)

Taylor relocated and re-described the same boulders and several others from the Fremington and Saunton clays.

1. No. 6 – Crinan spilite
• The same as Dewey’s Boulder III.
• Suggested similarity to Crinan spilitic pillow lavas (Cornwall).
• Probable Cornish source.
2. No. 7 – Hyalopilitic andesite
• Dewey’s Boulder IV.
• Glassy andesite, well-rounded, no augite.
• Similar to Dumfries and Loch Craignish rocks, but not identical.
• Possible Scottish or western British source.
3. No. 8 – Quartz porphyry (Fishley Pottery pit)
• Light grey, granitic texture, much-altered quartz and felspar.
• Possible local source, perhaps a Devon or Cornwall porphyritic dyke.
4. Nos. 9 & 10 – Quartz dolerite and olivine dolerite (Brannam’s clay pits)
• Compact igneous rocks, rounded or angular, unstratified in the Fremington Clay.
• No diagnostic features for distant transport.
• Suggested local origin, possibly Devon intrusions.

Original Provenance Views

• Dewey: Cautious; local/Cornish likely, Scottish comparisons tentative.
• Taylor: Strongly local (Devon–Cornwall intrusions/dykes); erratics support periglacial/fluvial reworking, not distant ice.

Modern Interpretation (Post-1956)

1. Nature of the Fremington Clays

Modern sedimentological and palaeoclimatic work (Croot et al., 1996; Bennett et al., 2024); interprets the Fremington Clays as:

• Late Pleistocene slope and periglacial deposits, possibly including solifluction and fluvial components.
• Not definitively glacial till — no evidence for direct ice-sheet deposition in North Devon.
• Erratics were likely incorporated into the clay by local mass-movement or river reworking, not transported by ice.

2. Source and Petrology

Later petrographic and geochemical comparisons confirm:

• These erratics from the Fremington/Bickington clays (near Combrew Farm) could plausibly all derive from the northern Dartmoor aureole (e.g., Meldon-Sourton inliers, SX 56-53), a Carboniferous volcanic-sedimentary sequence with spilites, andesites, and intrusives partially hornfelsed by the granite. Fluvial/glacial transport via the Taw/Okement system (10–20 km) fits, as BGS maps show matching lithologies in the Posidonia Shale and Crackington Formations. Key alignments:

  • Boulders III/6 (spilite-like): Albite-rich, chloritised, vesicular with secondary granophyric infill—mirrors Meldon spilitic pillow lavas (chlorite-epidote altered basalts); granophyric is a low-grade metamorphic overprint, not requiring Cornish import.
  • Boulder IV/7 (hypersthene andesite): Porphyritic acid labradorite, exclusive rhombic pyroxene (hypersthene), glassy base, no augite—directly matches rare hypersthene-phyric andesite lenses in Meldon tuffs, distinct from Scottish augite-bearing types.
  • Boulder 8 (quartz porphyry): Altered plagioclase-quartz with epidote, apatite prisms—resembles porphyritic rhyolitic dykes/sills in the aureole's eastern margin (e.g., near Mary Tavy), where devonisation yields similar amorphous matrices.
  • Boulders 9/10 (quartz/olivine dolerites): Fine-grained, ophitic with augite/olivine, micropegmatite, ilmenite—common in aureole basic intrusives (e.g., Meldon dolerite sheets), including kaolinised felspars from hydrothermal alteration; apatite needles are diagnostic of local Carboniferous mafics.
  • Minor finds (olivine dolerite pebble, Carboniferous grit): Ubiquitous in aureole stream gravels; oxide skins suggest pre-depositional wear, aligning with Okement-derived entrainment.

  While some (e.g., porphyry) allow non-local alternatives (Cornish dykes), the cluster's mafic-intermediate dominance and shared alteration (chlorite, epidote, calcite) strongly favour a unified aureole source over disparate exotics.

Boulder	Lithology	Key Features	Original Source Suggestion	Modern Consensus
III/No. 6	Spilite	Vesicular albite, micropegmatite, chlorite	Cornwall spilites	SW England volcanics (Cornwall/Dartmoor)
IV/No. 7	Hypersthene andesite	Glassy labradorite, hypersthene, no augite	Scottish (tentative)	Local SW dykes (Tamar/Dartmoor)
No. 8	Quartz porphyry	Altered phenocrysts, epidote	Local dyke	Devon–Cornwall intrusions
No. 9	Quartz dolerite	Ophitic augite, apatite	Local Devon	Dartmoor dykes (Meldon)
No. 10	Olivine dolerite	Yellow olivine, ilmenite	Local Devon	Devon minor intrusions

 

Conclusion: Erratics are local, fitting periglacial/fluvial processes; no glacial long-distance transport needed.

No modern studies have found evidence of northern or foreign erratics in the Fremington Clays. Their presence aligns with local volcanic and intrusive lithologies of Devon and Cornwall, reworked into periglacial or fluvial clays during Late Pleistocene climatic episodes.

 

 

 

The Stanton Drew Sarsen?

 Hautville’s Quoit

 

Hautville's Quoit – Historic England Photo - https://historicengland.org.uk/listing/the-list/list-entry/1002475

Abstract

Hautville’s Quoit, the recumbent megalith northeast of the Stanton Drew stone circles in Somerset, has traditionally been identified as a sarsen—a silica-cemented sandstone akin to those at Avebury and Stonehenge. Petrographic work by the Bath and Camerton Archaeological Society (Richards et al, BACAS, 2012) offers a more nuanced picture: while petrographic and macroscopic attributes broadly match sarsen lithology, the presence of small fossil impressions within the stone complicates this identification. The results suggest that the “Stanton Drew Sarsen” may represent a fossiliferous variant of silcrete derived from the same Paleogene strata that produced the Wiltshire sarsen field, rather than a local sandstone or glacial erratic.

---

1. Introduction

The Stanton Drew complex, comprising three major circles, avenues, and outliers, ranks among Britain’s most significant Neolithic monuments. Its lithological variety—dolomitic conglomerates from the Harptree area and oolitic limestones from Dundry (Richards et al. 2012)—implies deliberate selection reflecting local landscape and ritual integration. Hautville’s Quoit, situated c. 400 m northeast of the Great Circle across the River Chew, stands apart: a pale brown-grey sandstone slab, recumbent but once upright, visually aligned with the circles in a processional axis (Mercer 1969).

Earlier commentators from Stukeley to Lloyd Morgan (1887) debated its identity, proposing origins from Wiltshire’s Marlborough Downs. Modern analytical work upholds a sarsen-like composition yet introduces a critical complication—its embedded bivalve-like fossils, previously undocumented in Wiltshire material.

---

2. Lithological and Petrographic Characteristics

BACAS Field microscopy (×30) reveals a fine- to medium-grained (250–375 µm) quartz sandstone: well- to medium-sorted, subrounded to rounded grains of high sphericity, bound in siliceous (opal/chalcedony) cement yielding a shiny, translucent lustre (Richards et al. 2012). The matrix-supported fabric includes pock-marks (rootlet dissolution?), striations (polygenetic weathering), and lichen-scaled patches. Crucially, bedding planes expose small bivalve fossils (up to 10 mm × 6 mm)—clam-like, decalcified imprints—absent in local Triassic Mercia Mudstone or Jurassic equivalents. Dimensions (visible: 2.1 × 1.4 × 0.6 m) imply an original >30-ton block, reduced by 18th-century quarrying.

---

3. Provenance and Comparative Analysis

BACAS (2012) conducted systematic geological comparison against regional lithologies. Candidate sources included:

• Mendip quartzitic sandstones and Carboniferous grits – angular grains and ferruginous cement inconsistent with the Quoit’s fabric;
• Upper Greensand cherts of the Blackdown Hills – fine-grained, fossiliferous but lacking the Quoit’s coarse quartz matrix;
• Coal Measure ganisters near Pensford – similar cement but unsuitable texture;
• Somerset silcretes (South Petherton, Chew Valley) – non-fossiliferous and visually distinct;
• Wiltshire sarsens (Fyfield Down, Marlborough Downs) – close petrographic and macroscopic match, with similar grain size, sorting, silica cement, and colour.

The Wiltshire correlation remains the strongest, and the authors favour this provenance, citing similarities in surface polish and weathering patterns to the West Kennet Avenue stones at Avebury (Mercer 1969; Richards et al. 2012).

The absence of glacial erratic indicators further supports deliberate transport. However, the fossil content remains problematic: Wiltshire sarsens generally lack biogenic traces due to total silicification. BACAS hypothesised that Hautville’s Quoit originated from a fossiliferous Paleogene lamina within the Reading or Lambeth Group—possibly an outlying or peripheral depositional pocket not represented in sampled silcrete cores.

Comparable yet non-identical stonework includes the Pool Farm cist slab at West Harptree, a Bronze Age fossiliferous sandstone exhibiting corbulid and nuculid voids (Grinsell & Taylor 1956; Coles et al. 2000). XRF data, however, preclude direct correlation.

---

4. The Fossil Problem

A distinctive feature of Hautville’s Quoit is the reported presence of small bivalve-like fossil impressions. These were recognised during microscopic inspection but not taxonomically identified. Their occurrence is anomalous: true sarsen silcretes typically lack macrofossils because the silicification process obliterates organic material.

BACAS proposed that the fossils might represent relict shells incorporated into the original Paleogene sand beds prior to silicification, perhaps from a localized shelly lamina within the Reading or Lambeth Group. The lack of equivalent fossils in surveyed Wiltshire sarsens suggests either (a) the Quoit derived from a specific fossiliferous facies within the same formation, or (b) that it represents a silicified sandstone variant distinct from classic “grey wethers” of the Marlborough Downs.

 

---

5. Archaeological and Symbolic Context

The likely long-distance transport of such a massive stone—from the Marlborough or Pewsey Downs, over 50 km as the crow flies—would underscore Stanton Drew’s participation in a shared monumental tradition with Wessex. However, if the stone proves to be a fossiliferous local sandstone, its procurement would represent a regional adaptation of that tradition using available materials.

Either outcome alters the narrative: the “Stanton Drew Sarsen” is no mere erratic, but a deliberately selected and possibly symbolically charged lithology, chosen for its visual or tactile qualities and placed in alignment with the circles across the river valley.

---

6. Conclusion

Hautville’s Quoit remains one of the most enigmatic stones in the West Country. The BACAS study demonstrates that its lithology most closely resembles sarsen from the Marlborough Downs, yet its fossil content defies a simple classification. Whether a rare fossiliferous silcrete or an anomalous sandstone, the “Stanton Drew Sarsen” bridges geological and cultural frontiers, linking Somerset’s great circle to the wider Neolithic megalithic tradition of southern Britain.

---

References

 Coles, J., Gestsdóttir, H., and Minnitt, S. 2000. A Bronze Age Decorated Cist from Pool Farm, West Harptree: New Analyses. Proceedings of the Somerset Archaeological and Natural History Society. Volume 144

Linford, Neil & Linford, Paul & Payne, Andrew & Greaney, Susan. (2017). Stanton Drew Stone Circles and Avenues, Bath and North East Somerset, Report on Geophysical Surveys, July 2017. 10.13140/RG.2.2.33792.28165.

Lloyd-Morgan, C. 1887. The Stones of Stanton Drew: their source and origin, Part II. Proceedings of the Somerset Archaeological and Natural History Society. Volume 33 

Mercer, R. 1969. Hautville’s Quoit Excavation Notes includes Clark, A.J. Geophysical Survey Report (Unpublished)

Richards, J., & Oswin, J. Geophysical Survey at Stanton Drew, July 2009.

Richards, J., Oswin, J., and Simmonds, V. 2012. Hautville’s Quoit and other investigations at Stanton Drew. Bath and Camerton Archaeological Society in collaboration with Bath & North East Somerset Council.

Richards, J. (2023) ‘Pool Farm cist-slab, Hautville’s Quoit, and Fyfield Down Sarsens’, Rambling On [blog], 19 September. Available at: https://ramblingon.mendipgeoarch.net/2023/09/19/pool-farm-cist-slab-hautvilles-quoit-and-fyfield-down-sarsens/ (Accessed: 26 October 2025).

 

Prehistoric Landscape Change Around the Sources of Stonehenge’s Bluestones in Preseli, Wales

**Authors:** Daisy Eleanor Spencer, Karen Molloy, Mike Parker Pearson, Ralph Fyfe, Aaron Potito  

**Journal:** Landscape Research (published online ahead of print, 2025)  

**DOI:** [10.1080/14614103.2025.2574741](https://doi.org/10.1080/14614103.2025.2574741)  

This paper presents a palaeoenvironmental study examining landscape changes in the Preseli Hills of Pembrokeshire, west Wales—the source region for Stonehenge's iconic bluestones. The research tests the hypothesis that the transportation of these stones to Stonehenge around 3000 BC (marking the site's earliest phase) may have coincided with a significant out-migration of people from Preseli, potentially leading to reduced human activity until about 2200 BC. This could manifest in the palaeoecological record as a decline in indicators of human land use, such as pastoral or arable farming signals.

#### Methodology

The study employs a multi-proxy, multi-site approach to overcome the challenges of limited deep peat sequences in the region. Researchers analysed multiple sediment cores from heathland slopes in the Preseli Hills, supplemented by archaeological spot samples from five local prehistoric sites. Key analyses included:

- Pollen and macrofossil identification to track vegetation changes.

- Non-pollen palynomorphs (NPPs) for additional environmental indicators.

- Loss-on-ignition (LOI at 550°C) for organic content assessment.

- Stratigraphical profiling for chronological context.

This spans the Early Holocene to the Late Bronze Age, providing high-resolution insights into local environmental dynamics.

#### Main Findings

- **Mesolithic and Early Neolithic:** The landscape was overwhelmingly wooded, with dominant woodland species in pollen records, reflecting minimal human intervention.

- **Early–Middle Neolithic:** Subtle increases in pastoral indicators (e.g., grazing-related pollen) align with archaeological evidence of activity at key bluestone quarries and monuments, suggesting emerging human influence contemporaneous with Stonehenge's bluestone phase.

- **c. 3000–2200 BC (Beaker Period):** Despite sparse archaeological artefacts indicating potential depopulation, the presence of cereal pollen grains points to ongoing, albeit low-level, human presence and agricultural activity. No sharp decline in anthropogenic signals was detected, challenging the out-migration hypothesis.

- **Late Bronze Age:** A marked expansion of pastoral farming and arable cultivation occurred, with significant rises in grassland and crop-related pollen, indicating intensified land use.

The findings highlight continuity in human-environment interactions rather than abandonment post-bluestone transport.

#### Contributions and Implications

This work advances understanding of prehistoric population dynamics in Preseli by integrating palaeoecology with archaeology, filling gaps in a region where data scarcity has long hindered interpretations. It suggests that bluestone movement did not trigger mass exodus but coexisted with sustained, if modest, local habitation. The study underscores the value of multi-core sampling for robust reconstructions in peat-poor landscapes and contributes to broader debates on Neolithic–Bronze Age transitions in Britain, including cultural links between Wales and Wiltshire.

Caution in Attributing the Fremington Clay Series to Irish Sea Glaciation: A Case for Predominantly Fluvial and Periglacial Origins in North Devon

Caution in Attributing the Fremington Clay Series to Irish Sea Glaciation: A Case for Predominantly Fluvial and Periglacial Origins in North Devon

Abstract

The Fremington Clay Series of north Devon has been central to debates on the extent of Middle Pleistocene glaciation in south-west England, often interpreted as evidence of Irish Sea ice incursion during the Wolstonian Stage (MIS 6). However, stratigraphic, sedimentological, petrological, geomorphological, and chronological evidence, drawn from historical and recent studies, warrants caution in this attribution. This paper synthesises data from key exposures (e.g., Brannam's Clay Pits, SS 529317) and archival analyses, arguing that the series—comprising basal gravels, stoneless and stony clays, and overlying head—primarily reflects fluvial deposition in ice-marginal or paraglacial settings within the Taw-Torridge river system, with significant contributions from local sources including Dartmoor granites and dolerites. Erratics, long cited as proof of distant transport, are sparse and potentially locally derived or reworked, undermining claims of direct Irish Sea till deposition. This updated synthesis incorporates detailed records of erratics excavated directly from the Fremington Clay (e.g., Arber, 1964; Taylor, 1956; Dewey, 1910), highlighting their lithological diversity and affinity to Dartmoor aureole rocks, while distinguishing them from far-travelled, ice-rafted boulders documented on adjacent north Devon coasts (e.g., Saunton and Croyde Bay). Integrating insights from Croot et al. (1996), Wood (1974), and recent critiques (e.g., Daw, 2024a; 2025a; Bennett et al., 2024), we highlight ambiguities in provenance and age (potentially Anglian, MIS 12, per OSL), advocating reanalysis of archival clasts via modern geochemistry. This fluvial-periglacial model resolves depositional inconsistencies, confines onshore glaciation to thin, localised ice caps, and aligns with offshore Bristol Channel evidence, offering a parsimonious framework for Devon's Quaternary record.

Keywords: Fremington Clay, fluvial deposition, erratics, Dartmoor provenance, Pleistocene south-west England, Irish Sea glaciation

1. Introduction

The Quaternary glacial history of south-west England remains contentious, particularly regarding the southerly limits of Irish Sea ice sheets. While offshore evidence from the Celtic Sea and Western Approaches indicates a long-lived ice margin at approximately 51°N during multiple cold stages (Wingfield, 1995), onshore corroboration south of this latitude is sparse and disputed. The Fremington Clay Series, exposed along the Taw estuary near Barnstaple (e.g., Brannam's and Higher Gorse Clay Pits, SS 529317–530316), stands as the principal candidate for terrestrial glacial deposits in Devon (Stephens, 1970; Wood, 1974). Described since Maw (1864) as a compact, variably stony clay with exotic erratics, it has been interpreted as a Wolstonian till or glaciolacustrine sequence deposited by Irish Sea ice impinging on the north Devon coast (Stephens, 1966, 1970; Kidson & Wood, 1974).

This glacial paradigm, revived in the mid-20th century against earlier fluvial-lacustrine alternatives (Balchin, 1952; Prestwich, 1892), relies on stratigraphic superposition (overlying Devensian head, underlain by Hoxnian gravels), foraminiferal assemblages suggestive of marine reworking, and far-travelled erratics (e.g., Scottish granites). However, subsequent investigations, including a major 1994 excavation (Croot et al., 1996), have revealed horizontal bedding, diffuse contacts, and predominantly local clast provenances, favouring a glaciolacustrine or fluvial origin in an ice-dammed Taw valley. Recent syntheses further question high-level onshore glaciation, attributing erratic clusters to sea-ice rafting or fluvial transport (Daw, 2024a; 2025a; Bennett et al., 2024).

This paper argues for caution in designating the Fremington Clay Series as unequivocal evidence of Irish Sea glaciation. By integrating historical data (1974 QRA Handbook; Wood, 1974) with modern sedimentology (Croot et al., 1996) and emerging critiques, we propose a predominantly fluvial-periglacial model, emphasising local riverine deposition augmented by Dartmoor-sourced materials. This updated analysis incorporates primary records of erratics found in situ within the clay (Arber, 1964; Taylor, 1956; Dewey, 1910; Vachell, 1963), which reveal a suite of igneous lithologies compatible with Dartmoor aureole sources rather than distant ice transport. These are differentiated from the more abundant, far-travelled ice-rafted erratics on north Devon beaches (e.g., Saunton and Croyde; Taylor, 1956; Madgett & Inglis, 1987), which have often been conflated with inland finds. This interpretation addresses stratigraphic ambiguities, the minor role of exotics, and chronological discrepancies (e.g., OSL ages >424 ka BP), while confining glaciation to offshore realms and thin Dartmoor ice caps. We structure the discussion around historical interpretations (Section 2), stratigraphic-sedimentological evidence (Section 3), erratic provenance (Section 4), geomorphological context (Section 5), chronological constraints (Section 6), and broader implications (Section 7).

2. Historical Interpretation of the Fremington Clay Geology

The historical interpretation of the Fremington Clay in north Devon as a fluvial deposit emerged in the late nineteenth century, amid efforts to contextualise southern England's Pleistocene valley infills within a framework of post-glacial sea-level fluctuations rather than direct continental glaciation. Joseph Prestwich (1892) provided a foundational fluvial-estuarine model, describing the clay as an overbank accumulation in a river-fed embayment of the Taw estuary, characterised by fining-upward sequences from subangular local gravels to stoneless silts, overlain by pebbly sands indicative of episodic flood settling. This view, building on George Maw's (1864) initial mapping but rejecting his boulder-clay attribution, aligned the deposit with raised beaches at 15–20 m OD, such as those at Penhill Spit, and emphasised its confinement to the valley floor without evidence of widespread ice override. Complementing Prestwich, William Ussher (1878) interpreted the underlying gravels as Taw River alluvium, correlating them with estuarine terraces and highlighting the absence of exotic clasts or shear fabrics that might imply glacial transport. These early syntheses positioned the Fremington Clay as a product of oscillatory submergence and fluvial aggradation, resolving stratigraphic inconsistencies by invoking local sediment recycling over far-travelled ice-sheet debris.

Mid-twentieth-century reappraisals refined this fluvial paradigm, integrating periglacial influences while countering resurgent glacial hypotheses. Wilfrid Balchin (1952) reframed the clay as an alluvial infill of oxidised Keuper Marl in a periglacial floodplain, underscoring its red-brown matrix, homogeneous texture, and lateral pinch-out as signatures of terrestrial reworking rather than glaciomarine diamicton. George Mitchell (1960) acknowledged hybrid elements but prioritised fluvial origins for the basal units, interpreting scattered pebbles as flood-emplaced rather than ice-rafted. Edmund Edmonds (1972) further advanced a non-glacial model for the pebbly drifts at Fremington Quay, viewing them as solifluction reworked by Ipswichian floods into river terraces, with weak imbrication and grading to Hoxnian beaches precluding override. These interpretations, echoed in regional geomorphological surveys, challenged Frederick Zeuner's (1959) bottom-moraine proposal by emphasising paraglacial drainage diversions in the Taw-Torridge basin, thus confining Dartmoor-derived materials to braided-stream deposition during Anglian (MIS 12) cold phases.

However, this fluvial consensus was persistently muddied by confusion with far-travelled coastal erratics at sites like Croyde Bay and Saunton Sands, which early observers conflated with the clay's sparse embeds to bolster Irish Sea glaciation claims. Henry Dewey (1910, 1913) extended Maw's correlations, interpreting hypersthene-andesites and granophyres in the clay as akin to the exotic gneisses and porphyries (up to 50 tonnes) on raised platforms, suggesting unified ice transport despite stratigraphic disparities—the clay overlying equivalents of the 7.5 m OD Patella Beach. Charles Taylor's (1956) catalogues exacerbated this by grouping 'Saunton and Fremington erratics' indiscriminately, amplifying onshore ice narratives without distinguishing the coastal boulders' subrounded, striated forms clustered in head or beach gravels from the clay's subangular, aureole-affine pebbles at 2–22 ft depths. This lumping overlooked elevation mismatches and transport vectors, perpetuating 'myths' of high-level incursions.

Clarification emerged in the late twentieth century through targeted reappraisals that disentangled these suites via sedimentology and provenance. Everard et al. (in a 1960s raised-beach synthesis) explicitly refuted glacial linkages, noting: 'Fremington boulder clay overlies the equivalent of the Raised Beach, it cannot have been responsible for the coastal erratics found at Croyde and Saunton,' attributing the latter to ice-floe rafting during Wolstonian interstadials. Madgett and Inglis (1987) surveyed 37 Saunton-Croyde boulders, correcting Taylor's misidentifications and differentiating them as sea-ice proxies from the clay's solifluction terraces, with minor overlaps (e.g., reworked flints) as periglacial downslope lags. Modern syntheses, such as Harrison (1997) in the Geological Conservation Review and Bennett et al. (2024), reinforce this resolution, portraying the clay as a continuous 4 km fluvial body with pseudo-laminated fines, while coastal erratics reflect Celtic Sea calving—thus restoring a parsimonious fluvial-periglacial narrative for Devon's Quaternary record.

3. Stratigraphy and Sedimentology: Signatures of Fluvial Rather Than Glacial-Marine Deposition

The Fremington Clay Series, up to 30 m thick, overlies a sub-Cainozoic rock platform (Crackington Formation) and is capped by periglacial head. Croot et al. (1996) delineated five units from a 1994 trench excavation (50 m N-S, 20 m E-W), expanding on Wood's (1974) "twin tills + outwash" model (Table 1). Key features include horizontal to pseudo-laminated bedding, fining-upward trends, and weak fabrics, inconsistent with subglacial lodgement but compatible with low-energy fluvial or lacustrine settling.

Unit	Description	Thickness (m)	Key Sedimentological Features	Interpretation (Croot et al., 1996; Wood, 1974)
E (Head)	Gravelly sand/clay; angular local clasts in yellow-brown matrix.	1–1.5	Uniform; gradational base; cryoturbated.	Periglacial solifluction (Devensian+).
D	Clast-rich (>50%) weathered red clayey silt; small gravels akin to Unit A.	0.5–1.0	CaCO₃ 10–20%; ill-defined deformation; over-consolidated.	Weathered glaciolacustrine/fluvial clay; post-depositional oxidation.
C	Irregular sand/silt lenses (quartz, haematite, local clasts); reworked fossils.	2–2.5	Sharp contacts; no grading; OSL >26 ka BP (minimum).	Ice-proximal fluvial sands; episodic flood inputs.
B	Dark brown clay; stoneless base (5% clasts) fining to clast-rich top (40%).	8–9	Diffuse laminae; no fabrics; >1500 clasts (16–256 mm) analysed.	Low-energy overbank/lacustrine; upward-increasing dropstones or flood boulders.
A (Basal)	Clast-supported subangular gravels; sandy-silt matrix (70:30 clast:matrix).	1.5–2.0	Weak NW-SE imbrication; all local (Crackington Fm.); erosional base.	High-energy fluvial/proglacial outwash.

Table 1. Revised stratigraphy of the Fremington Clay Series (adapted from Croot et al., 1996; Wood, 1974).

The basal Unit A gravel, poorly sorted (median 3.4 mm) and angular, lacks the rounding of Hoxnian raised beaches (e.g., Penhill spit, SS 519330, undisturbed at 16.7 m OD; Kidson & Wood, 1974). Instead, its grading and interdigitation with clays suggest braided-river deposition from seasonal snowmelt in adjacent valleys (Edmonds, 1972). The overlying stoneless clay (Unit B base) exhibits pseudo-laminae and fining upwards, hallmarks of fluvial overbank fines rather than uniform till matrix (contra Stephens, 1966). Scattered pebbles in the stony upper clay indicate episodic floods, not ice-rafted dropstones, as fabrics are absent and clasts subangular (Croot et al., 1996).

Foraminifera (e.g., Ammonia beccarii, Nonion labradoricum; Haynes in Wood, 1974) are derived and damaged, compatible with fluvial reworking of coastal marine sediments rather than primary glaciomarine input (contra Eyles & McCabe, 1989). Micromorphology reveals no glaciotectonic shear, only post-depositional deformation from over-consolidation, attributable to ice-proximal loading without direct override (Croot et al., 1996). This aligns with Balchin's (1952) lacustrine proposal and Prestwich's (1892) river-fed lake model, reframed here as a paraglacial floodplain in the Taw-Torridge basin.

4. Provenance of Erratics: Local and Reworked Sources Over Distant Irish Sea Transport

Erratics have anchored glacial interpretations since Maw (1864), who correlated inland boulders with coastal examples at Croyde Bay. However, a critical distinction must be drawn between the sparse erratics documented in situ within the Fremington Clay itself (e.g., Arber, 1964; Taylor, 1956; Dewey, 1910) and the more numerous, far-travelled ice-rafted boulders on adjacent north Devon beaches (e.g., Saunton and Croyde Bay; Taylor, 1956; Madgett & Inglis, 1987). The latter, often lumped together in glacial models, include unambiguous Scottish and Irish Sea lithologies (e.g., Ailsa Craig microgranite, Purbeck flint) deposited via sea-ice rafting or storm transport during lowstands, but these are absent or rare inland. In contrast, the clay-embedded erratics—primarily igneous types excavated from depths of 2–22 ft within the clay—are dominated by local to regional sources, particularly from the Dartmoor aureole (e.g., Permian-Triassic dolerite dykes, Variscan granites), mobilised via fluvial or periglacial processes.

Petrological inventories of clay-embedded erratics (Table 2, updated with in situ records from Dewey and Taylor) list igneous/metamorphic types (dolerite, granophyre, andesite) amid dominant local Devonian-Carboniferous clasts (>99%; Croot et al., 1996). Dewey (1910) and Taylor (1956) provide detailed thin-section analyses, confirming igneous dominance (e.g., spilitic textures in No. 6, ophitic in No. 10) with local affinities, such as Cornish spilites or Devon dykes, while noting morphological resemblances to Scottish types without geochemical confirmation. For instance, quartz-dolerites and olivine-dolerites match Meldon Chert Formation dykes, while hypersthene-andesites and granophyres evoke Dartmoor elvans and aureole rocks, distinguishable from Irish Sea equivalents via mineralogy (e.g., titaniferous augite in alkali micro-dolerites; Gilbert, pers. comm. in Croot et al., 1996). No. 8, an overlooked altered quartz porphyry from Fishley, exemplifies potential aureole sourcing, with epidote and apatite evoking Variscan intrusions mobilised via Taw floods.

Erratic No. / Location	Lithology / Type	Description & Notes	Proposed Glacial Source	Alternative Local/Regional Source (e.g., Dartmoor Aureole)	Key References
6 (Combrew Farm/Bickington)	Spilite (vesicular granophyre)	40×30×25 in; dark grey, porphyritic albite felspars, micropegmatite groundmass, chlorite-replaced ferromagnesian, secondary granophyric vesicles with calcite; no striae. Isolated in middle of clay-bed.	Irish Sea (Scotland).	N. Cornwall spilites (Crinan pillow-lava type) or Dartmoor volcanics.	Dewey (1910); Taylor (1956); Arber (1964).
7 (Combrew Farm/Chilcotts)	Hypersthene-andesite (hyalopilitic)	16 in across; dark grey-green, glassy porphyritic acid labradorite (zoned, twinned), hypersthene prisms (pleochroic), magnetite gridiron in brown glass base; no augite/olivine. ~22 ft below surface, c. 1870.	Irish Sea (Dumfries/Argyll).	Dartmoor elvan intrusions or W. Devon dykes.	Dewey (1910); Taylor (1956); Arber (1964).
8 (Fishley Pottery, near Combrew)	Altered quartz porphyry	47×19×16 in; light grey, holocrystalline granitic texture, phenocrystic quartz/felspar (up to 5 mm); crushed plagioclase, apatite prisms, red amorphous matrix, epidote. From clay-pit.	N/A (local?).	Porphyritic dyke W. of Devon/Cornwall coasts; Dartmoor aureole.	Taylor (1956).
9 (Brannam's pits)	Quartz-dolerite	c. 300 lb, ellipsoidal; grey, fine-grained, kaolinized felspar laths, primary quartz, fresh reddish augite, apatite needles, magnetite/calcite. In middle of brown clay.	Irish Sea (Scotland).	Dartmoor Permian-Triassic dykes (Meldon).	Taylor (1956); Arber (1964).
10 (Brannam's pits)	Olivine-dolerite	c. 300 lb, angular; darker grey, micro-pegmatitic ophitic, yellow olivine, ilmenite prisms, plagioclase tabs, slight quartz orientation. In brown clay; common Devon type.	Irish Sea.	Dartmoor olivine-bearing intrusions.	Taylor (1956); Arber (1964).
(Brannam's, 17 ft depth)	Olivine-dolerite pebble & Carboniferous grit slab	2-in rounded pebble (as No. 10); 5×1.25 in slab with red ferric oxide skin along cracks (post-inclusion infiltration).	N/A.	Local fluvial rework (pre-embedding waterworn).	Taylor (1956).
13 (Brannam's pits, 1962)	Quartz-dolerite	10 ft from top of clay.	Irish Sea (Scotland).	Dartmoor dykes.	Taylor (1956); Vachell (1963); Arber (1964).
(Higher Gorse, Plymouth 1994)	Alkali micro-dolerite	Small striated boulder in main clay unit; plagioclase phenocrysts, titaniferous augite, vesicles.	Irish Sea.	Dartmoor micro-dolerite variants.	Croot et al. (1996).
(Pen Hill, Taw Estuary)	Trachy-andesite	Partially buried in beach/estuarine sand (not in situ in clay).	Irish Sea.	Regional andesitic flows; fluvial rework.	Croot et al. (1996).
(Arber 1964, post-1957)	Dolerite and granodiorite	Removed boulders, originally inside clay; later identified.	Irish Sea.	Dartmoor aureole dolerite/granodiorite.	Arber (1964); Wood (1973).
(General Fremington area)	Spilite, grey elvan, quartz/olivine dolerite	Multiple small pebbles (50+), embedded 5–11 ft above base or at top/base.	Irish Sea.	Dartmoor aureole (elvan, spilite-like volcanics).	Taylor (1956); Croot et al. (1996); Arber (1964).

Table 2. In situ erratics in the Fremington Clay Series: Lithologies, descriptions, and alternative provenances (updated with Dewey, 1910; Taylor, 1956 records; excludes coastal ice-rafted boulders).

Sparse exotics (<1% >1.5 cm) occur as subangular pebbles or rare striated cobbles (e.g., single microdolerite at 4 m depth; Croot et al., 1996), embedded at low elevations (10–26 m OD). Granites match Dartmoor's Carboniferous pluton, mobilisable via periglacial clitter slopes and Taw entrainment (Evans et al., 2012). Dolerites align with local intrusions, distinguishable from northern equivalents via U-Pb/Hf isotopes—untested on archives (e.g., >1500 clasts at Plymouth University; Taylor's thin-sections at Cambridge). Flints and schorlrocks suggest short-distance fluvial/marine reworking, not ice-sheet transport (Daw, 2024a). Recent syntheses and petrological reappraisals continue to support a predominantly local or regional provenance, with the Dartmoor pluton and its aureole emerging as the most parsimonious source (Bennett et al., 2024). Even for enigmatic types like No. 6 and No. 7, Dartmoor affinities remain viable, with geochemical tracers recommended for confirmation (Daw, 2025a).

This profile favours hybrid fluvial-periglacial input: Taw-Torridge floods exported Dartmoor debris alongside local slates, explaining weak NW-SE fabrics without Irish Sea signatures. Rarity of true exotics (no chalk, minimal Scottish gneiss) and lack of concentration gradients refute sheet glaciation (Bennett et al., 2024), particularly when coastal rafted erratics are excluded.

5. Geomorphological Context: Ice-Marginal Rivers and Dartmoor Influence

The Taw-Torridge landscape evidences fluvial dominance. Wolstonian ice diversion blocked the estuary, reversing flow westward and damming a ~30 m OD lake graded to the third terrace (Edmonds, 1972; Stephens, 1966). This braided system, fed by Dartmoor meltwaters, deposited the Clay Series in a subsiding basin (Fig. 1, conceptual). Subtle Dartmoor moraines (Slipper Stones; Evans et al., 2012) imply thin, cold-based ice (<50 m thick), enhancing tors and dry valleys via frost action rather than erosion (Ballantyne & Harris, 1994).

Inland confinement (valley floors, 24 m deep at Roundswell) and absence of coastal drapes contradict glaciomarine models (Eyles & McCabe, 1989; contra Lambeck, 1995 sea-level constraints). Offshore tills correlate via palaeo-channels, but onshore, periglacial head and terraces prevail.

6. Chronological Constraints: Pre-Wolstonian Ages and Model Conflicts

OSL dating places Units B–C at >424 ka BP (Anglian, MIS 12; Croot et al., 1996), predating Wolstonian correlations (Stephens, 1970) and refuting Late Devensian glaciomarine flooding (Bowen, 1994). This aligns with Hoxnian underlain gravels but challenges Irish Sea synchrony with Scilly/Trebetherick "tills" (local at Trebetherick; Wood, 1973; Devensian at Scilly; Scourse, 1991). Variability in terrace grading (four levels; Edmonds, 1972) suggests multiple cold phases, with Fremington as Anglian fluvial legacy reworked in Wolstonian.

7. Broader Implications and Recommendations for Future Research

Attributing Fremington to Irish Sea ice has inflated onshore limits, sustaining "myths" of high-level glaciation (Daw, 2024a; John, 2024). A fluvial model confines ice to Bristol Channel seas, resolves erratic transport for megaliths, and emphasises periglacial valley carving (Bennett et al., 2024). It negates claims like Baggy Point's epidiorite (~80 m OD; Madgett & Madgett, 1974), reframed as fluvial or sea-ice.

Future work: Geochemical provenance (U-Pb on granites/dolerites); cosmogenic dating of terraces; re-excavation for intact faunas. This cautionary stance prioritises parsimony, highlighting Devon's fluvial sensitivity.

References

Arber, M. A. (1964). Erratic boulders within the Fremington Clay of North Devon. Geological Magazine, 101(3), 282–283. https://doi.org/10.1017/S0016756800049517

Balchin, W.G.V. (1946) The geomorphology of the North Cornish coast. Transactions of the Royal Geological Society of Cornwall, 17, 317-44.

Ballantyne, C. K., & Harris, C. (1994). The periglaciation of Great Britain. Cambridge University Press.

Bennett, J. A., Cullingford, R. A., Gibbard, P. L., Hughes, P. D., & Murton, J. B. (2024). The Quaternary Geology of Devon. Proceedings of the Ussher Society, 15, 84-130. https://research.manchester.ac.uk/en/publications/the-quaternary-geology-of-devon

Bowen, D. Q. (1994). The Pleistocene succession of the Gower Peninsula. In N. Stephens (Ed.), The Pleistocene of the Gower Peninsula (QRA field guide). Quaternary Research Association.

Croot, D. G., Gilbert, A., Griffiths, J., & van der Meer, J. J. (1996). The character, age and depositional environments of the Fremington Clay Series, North Devon. Quaternary Newsletter, 80, 1–15. https://www.qra.org.uk/wp-content/uploads/2020/08/QN80_1-15_Croot_etal.pdf

Daw, T. (2024) www.sarsen.org:  https://www.sarsen.org/2024/.

Daw, T. (2025). www.sarsen.org:  https://www.sarsen.org/2025/.

Dewey, H. (1910). Notes on some igneous rocks from North Devon. Proceedings of the Geologists' Association, 21(4), 429–434. (No DOI available; archived via Wiley:

Edmonds, E. A., Whittaker, A., and Williams, B. J. 1985. Geology of the country around Ilfracombe and Barnstaple. Memoir British Geological Survey, Sheets 277 and 293 (England and Wales).British Geological Survey England and Wales, Natural Environment Research Council. HMSO for the Natural Environment Research Council.

Edmonds, M. (1972). Exeter Museums Archaeological Field Unit report: Terrace stratigraphy in the Taw valley. Exeter Museums Archaeological Field Unit

Eyles, N., & McCabe, A. M. (1989). The Late Devensian (<22,000 BP) Irish Sea Basin: Glaciomarine environments of deposition. In J. Dowdeswell & J. D. Scourse (Eds.), Glaciomarine environments: Processes and sediments (pp. 249–271). Geological Society. https://doi.org/10.1144/GSL.SP.1989.053.01.13

Flett, J. S., & Hill, J. B. (1912). The geology of the Lizard and Meneage (Memoir of the Geological Survey of England and Wales: Sheet 359). HMSO.

Gibbard, Philip & Hughes, Philip & Rolfe, Christopher. (2017). New insights into the Quaternary evolution of the Bristol Channel, UK. Journal of Quaternary Science. 32. 10.1002/jqs.2951.

John, B. S. (2024, June 28). Stonehenge and the Ice Age [Blog]. Brian Mountain-Man. https://brian-mountainman.blogspot.com/2024/06/

Kidson, C. and Wood, R. (1974) The Pleistocene stratigraphy of Barnstaple Bay. Proceedings of the Geologists' Association, 85, 223–37.

Lambeck, K. (1995). Late Devensian and Holocene shorelines of the British Isles and North Sea from models of glacio-hydro-isostatic rebound. Journal of the Geological Society, 152(3), 437–448. https://doi.org/10.1144/gsjgs.152.3.0437

Madgett, P. A., & Inglis, T. H. (1987). A re-appraisal of the erratic suite of the Saunton and Croyde areas, North Devon. Transactions of the Devonshire Association. vol.119, pp 135-144.

Madgett, P.A. and Madgett, R.A. 1974. A giant erratic on Baggy Point, North Devon. Quaternary Newsletter, 14, 1-2.

Maw, G. (1864) On a supposed deposit of boulder-clay in North Devon. Quarterly Journal of the Geological Society of London, 20, 445–51

Prestwich, J. (1892) The raised beaches and 'head' or rubble drift of the south of England: and their relation to the valley drifts and to the glacial period; and on a late post-glacial submergence. Quarterly Journal of the Geological Society of London, 48, 263–343.

Scourse, J. D. (1991) Late Pleistocene Stratigraphy and Palaeobotany of the Isles of Scilly, Philosophical Transactions of the Royal Society of London Series B, vol. 334, no. 1271, pp. 405–448, 1991

Stephens, N. (1966). Some Pleistocene deposits in North Devon. Biuletyn. Peryglacjalny, 15, 103-114.

Stephens, N. (1970). The West Country and Southern Ireland.  In C. A. Lewis (Ed.), The glaciations of Wales and adjoining regions (pp. 2567-314). Longman.

Stephens, N. (1974) Some aspects of the Quaternary of South-West England; Westward Ho!; The Fremington area; North Devon; Hartland Quay and Datnhole Point; Chard area and the Axe Valley section. In Exeter Field Meeting, Easter 1974 (ed. A. Straw), Quaternary Research Association Field Handbook, Exeter, pp. 5–7; 25–7; 28–9; 35–42; 45; and 46–51.

Taylor, C.W. (1956) Erratics of the Saunton and Fremington areas. Report and Transactions of the Devonshire Associaton for the Advancement of Science, Literature and Art, 88, 52–64.)

Wingfield, R. T. R. (1995). A model of sea-level in the Irish and Celtic Seas during the end-Pleistocene to Holocene transition. Geological Society, London, Special Publications. 96. 209-242. 10.1144/GSL.SP.1995.096.01.15.

Wood, T. R. (1974). Quaternary deposits around Fremington. In A. Straw (Ed.), QRA Easter field meeting handbook (pp. 30–34). Quaternary Research Association. https://www.qra.org.uk/field_guide_archive/